Development of Indian Agricultural Research Ontology: Semantic Rich Relations Based Information Retrieval System for Vid
Digital Libraries represent semantically rich collections of digital documents. Ontology-based information retrieval systems capture semantic relations for providing value added information services. Deviating from the regular approach of developing ontol
- PDF / 403,309 Bytes
- 10 Pages / 430 x 660 pts Page_size
- 110 Downloads / 189 Views
Abstract. Digital Libraries represent semantically rich collections of digital documents. Ontology-based information retrieval systems capture semantic relations for providing value added information services. Deviating from the regular approach of developing ontologies on the basis of domain knowledge, the present paper puts forward a novel method for developing ontologies from the semantic information available in the titles of digital documents. Such an approach gathers significance due to its simplicity in ontology development process. To examine the same, the study considered the case of Agricultural Electronic Theses and Dissertations (ETDs) present in Vidyanidhi Digital Library. The study resulted in the development of Indian Agricultural Research domain ontology, which was used for developing ontology-based information retrieval system. This paper while describing the methodology followed for developing the ontology presents the technical details of the developed system. Keywords: Indian Agricultural Research Ontology, Ontology, Web Ontology Language, Semantic Web, Vidyanidhi Digital Library.
1 Introduction The field of Information Retrieval is a central area of research in Digital Libraries. Information Retrieval (IR) is a process of finding all relevant documents from a document collection, satisfying user information need [1]. Unfortunately, currently employed information retrieval mechanisms suffer from various limitations. Issues such as information overload, rapid technological developments, fluctuating user trends and behaviour call for better IR mechanisms. The emerging Semantic Web technologies such as ontologies promise knowledge-based systems capable of performing crucial tasks of information retrieval and extraction [2]. Domain ontology based systems supporting navigation and querying facilities form ideal information retrieval systems for digital libraries. The reasoning and querying capabilities offer valuable search strategies for digital libraries. Ontology-based IR systems mainly rely on domain ontologies, which are further extended for information retrieval. Development of domain ontologies is not only costly [3], time consuming and cumbersome, but also leads to difficulties particularly D.H.-L. Goh et al. (Eds.): ICADL 2007, LNCS 4822, pp. 400–409, 2007. © Springer-Verlag Berlin Heidelberg 2007
Development of Indian Agricultural Research Ontology
401
at the time of mapping document instances to the ontology. Further, the high volume of knowledge represented in the ontology may not be used in its entirety. Thus, instead of representing the entire knowledge into domain ontology and mapping documents to the ontology, it would be appropriate to develop ontology, based on the available documents and deploy the same for information retrieval. This would also facilitate in the easy mapping of documents to the ontology. The present paper puts forward a simple yet powerful method for developing ontologies from the information present in the titles of electronic documents, resulting in effecti
Data Loading...
